PROJECT SUMMARY Immune-mediated adverse drug reactions (IM-ADRs) are a major obstacle to the successful treatment of both HIV and tuberculosis (TB) internationally. Their contribution to management complexity and economic burden is a particular problem in South Africa (SA) where 1 in 4 individuals in the population is HIV-infected and 1 in 5 patients with HIV develops a cutaneous adverse drug reaction during treatment. Stevens-Johnson syndrome and toxic epidermal necrolysis (SJS/TEN) and drug reaction with eosinophilia and systemic symptoms (DRESS) are severe IM-ADRs that have mortalities that can exceed 40% and lead to prolonged hospitalization, higher healthcare costs and significantly constricted treatment options. The specific effect that these IM-ADRs have on HIV treatment outcomes has not been adequately measured. In addition, the IM-ADRs themselves are known to be associated with significant long-term disability and physical and mental health complications that have not been measured in HIV and HIV/TB co-infected populations. Preventive efforts for severe IM-ADRs such as DRESS and SJS/TEN have been fueled by promising discoveries such as the strong association between the HLA class I allele HLA-B*15:02 and carbamazepine SJS/TEN which has led to pre-treatment screening for HLA- B*15:02 before carbamazepine prescription in some Southeast Asian countries. Members of our group were responsible for the translation of HLA-B*57:01 screening to prevent abacavir hypersensitivity and we recently described a strong association between HLA-C*04:01 and nevirapine SJS/TEN in SA suggesting that HLA class I associations are also important for drugs used in the treatment and prevention of HIV. However, the genetic risk factors for IM-ADRs in African HIV-infected populations are incompletely understood and there are significant gaps in understanding the risk factors for IM-ADRs in drug commonly used in African HIV-TB co-infected populations. In SA and other resource poor African settings, given the high burden of HIV and TB, there is an urgent need to identify management strategies for IM-ADRs that will help improve prevention efforts, earlier diagnosis and treatment protocols. Our hypothesis tested will be that we will identify HLA and other genetic associations between SJS/TEN as well as DRESS and drugs used to prevent, treat and manage HIV and its co- morbidities. In specific aim 1 we will create a biorepository of DNA and other samples from IM-ADR cases related to drugs used to treat HIV and TB that includes underserviced areas in South Africa. Existing and new IM-ADR cases will undergo specific phenotype validation and causality adjudication. In specific aim 2 we will define HLA and other genetic risk factors associated with IM-ADRs in HIV/TB endemic settings. In specific aim 3 we will determine short and long-term complications and outcomes amongst a cohort of patients who have experienced IM-ADR and the specific impacts on HIV care. Using the synergistic gain of an existing US-South African collaboration we predict that our discoveries will create a roadmap for the prevention and management of IM-ADRs in complex HIV populations in resource poor settings internationally.